I’ve lived in the city of Boston for 17 years now - we moved here when my eldest was 9 months old. Along the way, I’ve watched many of our friends with kids of similar age leave Boston for the suburbs, with a primary reason cited being “for the schools.” I understand and share the desire for my kids to have access to a great education. But, if I am being honest, I think when we cite the “good schools” in suburbs, it’s really a euphemism for “schools where most of families have money and the kids are predictably similar to mine.” My wife and I made the choice to stay in Boston, because part of a great education is for our kids to grow up in a dynamic and diverse environment. But we also wanted them to go to an academically rigorous school with great teachers and challenging courses.

Boston Latin School is one of the rare places that can offer both. Founded in 1635, it is the oldest public school in the country. Its alumni roster includes John Hancock, Samuel Adams, Benjamin Franklin (though he never actually graduated), Ralph Waldo Emerson…the list goes on. It also has an admissions process, which, for decades, it has been a political lightning rod in terms of the policy for deciding who gets in. It is the epitome of a quintessentially American question: how do we balance merit and equity?

Two Bad Words…?

“Merit” and “equity” have, unfortunately, become politically radioactive terms. To the right, “equity” is code for performative quotas and reverse racism. To the left, “merit” is code for the preservation of systemic privilege rooted in historical discrimination. This kind of binary absolutism is unfortunate, because a healthy tension between these ideals is at the heart of the American project. Jefferson wrote to Adams in 1813 that there is “a natural aristocracy among men…grounded in virtue and talent”, and that the job of a good society is to distinguish it from the “artificial aristocracy founded on wealth and birth.” Two hundred years later, we’re still working on it. Over two hundred years later, Ronald Reagan said ““America’s greatest gift has always been freedom and equality of opportunity — the idea that no matter who you are, no matter where you came from, you can climb as high as your own God-given talents will take you.” Indeed, if “merit” ignores systemic barriers, it becomes a gatekeeper for privilege. If “equity” is pursued without rigor, it creates new forms of unfairness.

Balancing these two ideals is particularly tricky in the context of the admissions policy to an exam school in seventh grade, where students are right in the middle of their educational journey. When kids are just starting to learn, the concept of preschool admissions based on “merit” feels absurd (though it does ostensibly happen at elite private schools — can the four-year-old color inside the lines? In practice, I suspect these schools are mostly admitting based on parental wealth, but I digress). At the tail end of the educational path, when you’re selecting candidates for a surgical residency after 20+ years of schooling, merit should be carrying most of the weight, and ideally society has provided the opportunity-appropriate scaffolding along the way – such as a thoughtful 7th grade exam school admissions policy. The applicants to exam schools are kids who’ve had enough schooling that academic performance means something, but not so much that the system has had time to compensate for the significant resource gaps between families of varying means across the city.

The “Crisis” as a Catalyst

Before the pandemic, Boston used a single citywide pool for exam school admissions: a ranked list of all city students, weighted 50% grades and 50% exam score. Pure merit, zero equity, and the incoming class at BLS didn’t look much like Boston as a whole. An entire cottage industry had sprung up to help (mostly wealthier) families prepare (by paying for prep programs) for the ISEE exam. The test was administered on a Saturday and not aligned with the public school curriculum.

When COVID-19 hit and health restrictions prevented administration of the exam, BPS followed the old political adage: never let a good crisis go to waste. They scrapped the old system and a “task force” proposed a new model, ostensibly to create a more balanced and diverse student body. The intent was sound. The execution was not.

The new policy split the city into eight socioeconomic tiers using five census data variables. Each tier was allocated seats based on the number of potential (not actual) applicants. A new test, the MAP Growth, was introduced. And most confoundingly, a ten point bonus (on a one-hundred point school) was awarded to any applicant attending a Title I school (any school where more than 40% of students receive government assistance), regardless of which neighborhood that student lived in.

The Math Didn’t Add Up

On the margins, the policy produced absurd results. Several schools fell just barely on either side of the Title I threshold, with 39% or 41% of the students in the entire school qualifying. A single second-grader transferring in or out could tip a school across the line, determining whether any of its sixth-graders would have a realistic shot at Boston Latin School. (Not the kind of high-stakes math anyone wants a seven-year-old responsible for.)

After the first admissions cycle under the new system, the anecdotes were that it “was impossible” to get into BLS from a “Tier 8” neighborhood. Curiously, I sat down with the publicly available BPS data to try and understand morea about how the policy was actually performing. And it turned out the anecdotes were correct.

In “Tier 7” (the second most affluent socioeconomic tier), the ten-point school bonus had created a mathematical ceiling. For a high-performing student at a non-Title I school, getting into BLS required a composite score above 100. That’s not a figure of speech; it was a literal impossibility. The policy hadn’t leveled the playing field. It had tilted it so far in the one direction that it produced a new form of exclusion. It didn’t make it hard for these students to get into BLS - it made it impossible. And because the bonus was applied within a Tier, it was only impacting the competition between students in that socioeconomic tier, based on the school they attended. The bonus points did not have any impact on the socioeconomic diversity of the school.

I agreed with the hypothesis that, even when controlling for Tier, there may be an impact based on the nature of the school a student attended. So I analyzed the numbers and determined that the “natural” score gap, when controlling for Tiers, was just under two points on the hundred points scale. When I later spoke with people involved in drafting the original task force policy, I asked how they had landed on the number ten. The answer: it was round. Not derived from data. Not modeled against outcomes. Just a round number that sounded reasonable to someone in a room. The institution responsible for teaching Boston’s children to use evidence and reasoning had simply declined to apply either to this particular problem. And I (a random guy crunching numbers with no stake in the development of this policy) seemed to be the first person to publicly identify the absurdity.

A Warm Night in October

I wrote up my findings in a memo that included my analysis and proposed an update to the policy to use the “natural” raw score differentials for each tier rather than the arbitrary, and clearly unfair, value of 10. I signed up to present a public comment at a public Boston School Committee meeting. On a warm fall evening, I rode my bike downtown and walked into a large auditorium full of BPS administrators and several dozen public commenters.

I’ll be honest: I didn’t go in expecting much. I had no experience in politics at any level, and my assumption was that I’d get some polite nods and no perceptible change. I was confident in my analysis and the logic behind my propasal, but politics, I assumed, would matter more than math.

My name was called. I walked to the microphone, ran through the numbers as plainly as I could, made my pitch for change, and sat back down. The meeting moved on to the next speaker.

A few minutes later, I felt a tap on my shoulder. It was a committee member who asked if I could step in the hallway to talk with him. “I really like your idea,” he said. “It makes a lot of sense.”

He asked for my number. I gave it to him, still a little disoriented. I rode home that night feeling something I hadn’t anticipated: like it had actually mattered.

Over the following months, I talked several times with members of the school committee and BPS data team. The word from the school committee champion is that it was garnering support of the mayor (who chooses the school committee) and the superintendent, as the “need to score over 100” had become a thorny local news issue for both. In early 2023, the School Committee voted to adopt what was essentially my proposal.

On Iteration and Public Policy

One of the points I made in my memo was that the world I come from - software development - you iterate based on data. Modern software products are not in multi-year release cycles. We ship regularly based on new information and business needs. The stereotype is that government is not so nimble. But this experience demonstrated that doesn’t need to be the case. And BPS kept on listening and analyzing data in the coming years. After the initial policy change, BPS continued to present admissions data publicly after each cycle, and parents and advocates from across the political spectrum continued to show up and push for refinement.

That kind of sustained engagement matters, and it isn’t easy. Every policy change creates ripple effects. Families plan their lives around these systems, and the risk of unintended consequences is real. Changing course requires institutional humility that is hard to find in any governing body. But the School Committee proved willing to follow the data, even when that meant acknowledging that an earlier decision needed revision. After a few more admissions cycles, they made another round of changes that I believe has finally struck the right balance:

  • The first 20% of seats at each exam school are now awarded to the highest-ranking students citywide, regardless of where they live. This preserves the merit core of these institutions.
  • The remaining seats are distributed across four tiers (simplified from eight), determined by census tracts and five census data variables.
  • No more bonus points. Three years of data showed that the tiered system itself provided a sufficient equity lever, without additional complexity

So What?

I’m not a politician. I had no connections at City Hall. I just did the math, wrote it up, and went to a meeting to give my pitch. And it worked. Cynicism about politics is rampant. But this gives me hope. And it also gives me the confidence and motivation to throw myself into a much bigger arena that is similarly rooted in finding the right balance between equity / merit balance: the federal tax code!

The tax code has many of the same problems: arbitrary numbers nobody can justify, complexity that benefits the people who can afford to navigate it, and a system that hasn’t been seriously redesigned in decades. It’s a harder problem by orders of magnitude. But the approach is the same: start with the data, propose something specific, and see what happens.